计算机工程与应用Issue(19)：194-199,6.DOI:10.3778/j.issn.1002-8331.1309-0159
基于灰度-梯度二维对称Tsallis交叉熵的阈值分割
Thresholding segmentation based on gray-gradient 2-D symmetric Tsallis cross entropy
摘要
Abstract
2-D Tsallis cross entropy thresholding segmentation based on gray level-average gray level histogram exists misclassification, higher computational complexity. So thresholding segmentation based on gray-gradient 2-D symmetric Tsallis cross entropy is proposed. Firstly, a new 2-D histogram based on gray-gradient is created, and more fully considers the object points and background points. Then, the formulas of symmetric Tsallis cross entropy threshold selection based on the histogram zoning are derived. Finally, chaotic niche particle swarm optimization algorithm based on tent map is used to search for 2-D optimal threshold vector, and fast recursive algorithm is introduced to reduce the computational complexity of the fitness function. Experimental results show that this method enables the edge of the segmented image more accurate, more uniform gray within the class, and real-time increased by 30-fold, compared with 2-D Tsallis cross entropy segmentation based on gray level-average gray level histogram.关键词
阈值分割/对称Tsallis交叉熵/二维直方图/快速递推算法/混沌小生境粒子群Key words
threshold segmentation/symmetric Tsallis cross entropy/2-D histogram/fast recursive algorithm/chaotic niche particle swarm optimization分类
信息技术与安全科学引用本文复制引用
朱磊,吉峰,白瑞林..基于灰度-梯度二维对称Tsallis交叉熵的阈值分割[J].计算机工程与应用,2015,(19):194-199,6.基金项目
江苏高校优势学科建设工程资助项目（No.PAPD）；江苏省产学研前瞻性联合研究项目（No.BY2012056）。 （）
